Friday, November 4, 2011
The above graphs shows the weekly US oil consumption estimates from the EIA (blue - left scale), along with a 13 week centered moving average (red). This data is not zero-scaled to better show changes. The green line is oil prices (Brent - on the right scale). I have put vertical gridlines in for quarters as well as years.
If we look at the 2011 data, you can see that in the first half of the year the pattern was pretty much that oil prices were rising and consumption was falling (primarily as a result of the Arab Spring and the Libyan revolution). However, the last four months are a little harder to interpret. Clearly prices have been falling - if slowly - amidst the beginnings of economic contraction in Europe and fears over a wider financial crisis (with an assist from the debt limit debacle in the US).
The consumption trend in H2 is harder to call - you could say "roughly flat with some noise" - suggesting that the slowly falling prices have been enough to abate the drop in consumption. Alternatively, you could say it first went up at the end of Q2 and beginning of Q3 and then has been dropping in late Q3 and early Q4 (suggesting renewed economic weakness). I'm inclined to favor the first explanation but it's hard to say for sure.
The implication is for the level of oil prices. I have a rough rule of thumb that if US oil consumption is rising or even flat, then prices are below their long term future average: stagnant global production and rising consumption in Asia and the Middle East imply that US consumption must drop in the future. So depending on whether you interpret the drop in consumption in the last couple of months as noise or signal, oil prices are either too low, or not. I guess we'll have to come back to the question in a month or two.
(I should add - whether or not oil prices are below their long-term average now, there's certainly considerable potential for them to drop further depending on how much worse the European-centered debt problems get).